A method for eliminating cross-link interference using RIS under dynamic TDD

Abstract

The present invention discloses a method for eliminating cross-link interference by RIS assistance under dynamic TDD. In the scheme of the present invention, the reflection coefficient of the RIS element is designed by reconstructing the equivalent channel matrix. At the same time, it is considered to reduce the overhead by predicting the channel state information (Channel State Information, CSI) between the user and the RIS. The present invention can effectively alleviate the cross-link interference (Cross-link Interference, CLI), and under the same data rate, the present invention has better bit error rate (Bit Error Rate, BER) performance than the time division multiple access (Time Division Multiple Access, TDMA) scheme of the RIS assisted enhanced signal. When the CSI is obtained by prediction, the present invention can also maintain a low BER.

Description

Technical Field

[0001] The present invention belongs to the field of dynamic time division duplex (TDD) scenarios in the field of communications, and in particular to a method for eliminating cross-link interference by using a reconfigurable intelligent surface (RIS) under dynamic TDD. Background Art

[0002] One of the key features of the fifth-generation mobile communication network is the ultra-dense small cell deployment. The TDD-based air interface is proposed to be applied to the transmission of small cell signals, in which the uplink and downlink transmission resources of each time slot can be flexibly allocated. This TDD with flexible selection of uplink and downlink configuration is also called dynamic TDD. Dynamic TDD has high time resource utilization (some cells may not send or receive signals in the case of unified configuration), such as Hyejin Kim, Jintae Kim, Daesik Hong, "Dynamic TDD Systems for 5G and Beyond: A Survey of Cross-link Interference Mitigation," IEEE Communications Surveys & Tutorials, 2020. In the case of dynamic TDD flexibly configuring uplink and downlink transmission resources, each cell can adapt to business needs more flexibly, and this also has a certain effect on reducing base station energy consumption. However, the use of dynamic TDD will cause cross-link interference, which includes interference between base stations and interference between users, seriously affecting system performance. In order to solve the impact of cross-link interference, some researchers have proposed a method to mitigate cell cluster interference. This method can avoid strong cross-link interference within the same cluster, but traditional inter-cell interference still exists within the cluster. This method sacrifices system performance and cannot meet the traffic requirements of each cell. In addition, many researchers have proposed methods based on beamforming and interference alignment, but these methods require channel state information at the transmitter or require more feedback, which incurs considerable additional overhead. Some methods can only solve one of the interferences between base stations and between users. Summary of the invention

[0003] The present invention aims to provide a method for eliminating cross-link interference by RIS assistance under dynamic TDD to solve the technical problems in the above-mentioned background technology.

[0004] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0005] A method for eliminating cross-link interference by RIS under dynamic TDD, comprising the following steps:

[0006] Step 1: The transmitter forms a transmission signal; suppose the number of base station antennas is M, the number of user antennas is 1, the signal sent by the uplink cell user is x, and the signal sent by the downlink cell base station is y. Let each antenna of the base station send the same signal, that is, y = 1 M y, where 1 M represents an M×1 all-1 vector, y is the symbol sent by each antenna;

[0007] Step 2: Determine the minimum number of RIS components required. To ensure that interference can be completely eliminated, a sufficient number of RIS components are required. The number of RIS components N needs to satisfy:

[0008]

[0009] Where d BS1-BS2 represents the distance between base station 1 and base station 2, d BS1-RIS represents the distance between base station 1 and RIS, d BS2-RIS represents the distance between base station 2 and RIS, d UE1-UE2 represents the distance between user 1 and user 2, d UE1-RIS represents the distance between user 1 and RIS, d UE2-RIS represents the distance between user 2 and RIS, and the corresponding α represents the path loss exponent;

[0010] Step 3: The interference stack in the dynamic TDD system is represented as B, where each element of the vector B represents the interference received by the corresponding antenna; small-scale fading matrix information is obtained, where the small-scale fading matrix information of the user and RIS is obtained by prediction due to the large acquisition overhead; the equivalent channel matrix H is defined based on the obtained small-scale fading matrix information D , H D The physical meaning of the element in the mth row and nth column is the sum of all channel gains passing through the nth element of RIS to the mth receiving antenna;

[0011] Step 4: Reconstruct the channel matrix H D The reflection coefficient of RIS is determined by the interference B to the system Reflection coefficient of RIS in represents the amplitude coefficient of the nth element, θ n ∈[0,2π) represents the phase shift of the nth element; the reflection coefficient of RIS It is expressed as:

[0012]

[0013] in represents the inverse matrix.

[0014] Furthermore, the vector in step 3 is represented as: in is the small-scale fading matrix between base station 1 and base station 2; L BS1-BS2 is the path loss from BS1 to BS2, α BS1-BS2 is the path loss index between BS1 and BS2; is the small-scale fading matrix between UE1 and UE2; L UE1-UE2 is the path loss from UE1 to UE2.

[0015] Furthermore, in step 3, H D It is expressed as:

[0016]

[0017] in is the matrix H BS1-RIS The (i, j)th element of H BS1-RIS is the small-scale fading matrix between BS1 and RIS; is the matrix H BS2-RIS The (i, j)th element of H BS2-RIS is the small-scale fading matrix between BS2 and RIS; is the matrix H UE1-RIS The i-th element of H UE1-RIS is the small-scale fading matrix between UE1 and RIS; is the matrix H UE2-RIS The i-th element of H UE2-RIS is the small-scale fading matrix between UE2 and RIS; H D is a (M+1)×N matrix; H D The physical meaning of the element in the mth row and nth column of is the sum of all channel gains passing through the nth element of RIS to the mth receiving antenna.

[0018] By adopting the above scheme, the cross-link interference in the dynamic TDD system can be eliminated after the signal is reflected by the RIS, and at the same time, the signal can be enhanced to a certain extent, thereby improving the system performance.

[0019] The method for eliminating cross-link interference assisted by RIS under dynamic TDD of the present invention has the following advantages:

[0020] 1. The RIS-assisted method proposed in the present invention can eliminate both the inter-base station interference and the inter-user interference in the cross-link interference;

[0021] 2. The RIS-assisted method proposed in the present invention obtains the channel information between the user and the RIS by prediction, thereby reducing the overhead.

[0022] 3. The RIS-assisted method proposed in the present invention can enhance the signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of a system in which a cell interferes with an adjacent cell in a dynamic TDD system of the present invention;

[0024] Figure 2 It is a diagram of BER simulation results of the RIS-assisted scheme under dynamic TDD and the TDMA scheme of RIS-assisted enhanced signal proposed in the present invention at the same rate;

[0025] Figure 3 It is a simulation result diagram of theoretical BER and simulation value under the channel prediction condition of the present invention;

[0026] Figure 4 This is a diagram showing the influence of the CSI acquisition interval on the BER performance in the case of channel prediction of the present invention. DETAILED DESCRIPTION

[0027] In order to better understand the purpose, structure and function of the present invention, a method for RIS-assisted elimination of cross-link interference under dynamic TDD of the present invention is further described in detail below with reference to the accompanying drawings.

[0028] Figure 1 A two-cell dynamic TDD system model is given. Figure 1 As shown in the figure, in uplink transmission, edge user UE2 in cell 2 sends a signal to base station BS2. In downlink transmission, base station BS1 in cell 1 sends a signal to edge user UE 1. At this time, UE1 will receive interference from UE2, and BS2 will receive interference from BS1. Next, we place the RIS in the middle of the two cells and mitigate the impact of CLI by appropriately controlling the amplitude coefficient and phase shift of each RIS element.

[0029] Assume that base stations BS1 and BS2 are equipped with M antennas and the user is a single antenna. RIS consists of N elements. The small-scale fading matrices between BS1 and BS2, UE1 and UE2, BS1 and UE1, BS2 and UE2, BS1 and RIS, BS2 and RIS, UE1 and RIS, and UE2 and RIS are expressed as These matrix information are assumed to be known. The path loss from BS1 to BS2 is expressed as:

[0030]

[0031] Where d BS1-BS2 and α BS1-BS2 are the distance and path loss index between BS1 and BS2 respectively. UE1-UE2 , LBS1-UE1 , L BS2-UE2 , L BS1-RIS , L BS2-RIS , L UE1-RIS and L UE2-RIS The meanings are similar. Then the signal vectors received by BS2 and UE1 can be expressed as:

[0032]

[0033] Where v BS2 and v UE1 is complex Gaussian noise, and The signals sent by UE2 and BS1 respectively, is the reflection coefficient of RIS, in represents the amplitude coefficient of the nth element, θ n ∈[0,2π) represents the phase shift of the nth element.

[0034] Step 1: Let each antenna of the base station send the same signal, that is, y = 1 M y, where 1 M represents an M×1 all-1 vector. Therefore, all effective interference can be expressed as:

[0035]

[0036] Each element of vector B represents the sum of interference received by the corresponding antenna.

[0037] Step 2: Find the lower bound of N. The derivation assumes that the small-scale fading environment of the BS-UE, BS-RIS, and UE-RIS links is strong enough, where all elements in the fading matrix are 1. Then, in order to eliminate CLI, we need to satisfy that the power level of the reflected signal needs to be higher than the power level of the corresponding BS-UE link. Therefore, N needs to satisfy the following constraints:

[0038]

[0039] and

[0040]

[0041] That is, the number N of RIS components needs to satisfy:

[0042]

[0043] Step 3: The interference is stacked and represented as vector B, as shown in formula (3). Then the matrix H is defined by superimposing the channel gain D :

[0044]

[0045] in is the matrix H BS1-RIS The (i, j)th element of is the matrix H BS2-RIS The (i, j)th element of is the matrix H UE1-RIS The i-th element of is the matrix H UE2-RIS The i-th element of H. D is a (M+1)×N matrix. D The physical meaning of the element in the mth row and nth column of is the sum of all channel gains passing through the nth element of RIS to the mth receiving antenna.

[0046] Therefore, the signal vector received by BS2 and UE1 can be expressed as:

[0047]

[0048] in represents the Hadamard product.

[0049] Step 4 determines the reflection coefficient of RIS based on the reconstructed channel matrix and the interference suffered by the system. At the same time, in order to reduce the overhead of obtaining channel information, the channel information can be obtained through prediction.

[0050] Step 4.1 To eliminate cross-link interference (CLI), the RIS reflection coefficient The following constraints need to be met:

[0051]

[0052] Therefore, the reflection coefficient of RIS is as follows:

[0053]

[0054] in represents the inverse matrix.

[0055] Note that rank(H D )≤M+1, so when N≤M+1, there is no Therefore, the number of RIS elements N needs to satisfy formula (6).

[0056] After the RIS reflection coefficient is designed by the above method, the interference received by BS2 and UE1 after the signal is reflected by RIS can be eliminated, thereby reducing the impact of CLI on the dynamic TDD system.

[0057] Step 4.2: In practical applications, the cost of obtaining CSI between the user and the RIS is relatively large, so the cost is reduced by channel prediction, that is, the CSI between the user and the RIS is obtained and predicted regularly. Figure 1 In the scenario of UE1-RIS and H UE2-RIS The channel matrix H UE1-RIS The prediction of the lth value of the mth element in is based on the first k1 samples h UE1-RIS,m,l-1 ,...,h UE1-RIS,m,l-p :

[0058]

[0059] Where l2 = lmodk2, k2 is the time interval for performing channel estimation, k1 is the order of the AR model, and the optimal coefficient a j Determined by the orthogonality principle:

[0060] a=R -1 r (12)

[0061] Where a=(a1,…,a p ) T , R is the autocorrelation matrix, and the (i, j)th element of R r is the autocorrelation vector, the jth element of r Correlation coefficient R ij and r j It can be estimated from the observed samples without knowing the maximum Doppler shift or the number of scatterers in advance. The channel matrix H UE2-RIS The same method is used for prediction. Therefore, the matrix H of the prediction information is used D It is expressed as:

[0062]

[0063] The reflection coefficient of RIS can be expressed as:

[0064]

[0065] For UE1, the residual interference caused by the prediction error can be expressed as:

[0066]

[0067] Therefore, the residual interference average power is given by:

[0068]

[0069] The useful signal power received by UE1 is:

[0070]

[0071] Quadrature Amplitude Modulation (QAM) is used. Therefore, the BER of UE 1 can be calculated based on its modulation order I QAM =2 i express:

[0072]

[0073] where Q(.) is the Q function:

[0074]

[0075] The theoretical BER of BS can be expressed by the same process. UE1-RIS and H UE2-RIS is predicted, so there will be no residual interference at BS2, which will not affect its BER performance.

[0076] Figure 2 The BER simulation results of the proposed RIS-assisted interference elimination scheme under unpredicted conditions and the RIS-assisted enhanced signal under TDMA are given. Figure 3 The simulation results of theoretical BER and simulated values ​​are given under channel prediction. Figure 4 The influence of CSI acquisition interval on BER performance under channel prediction is given. In the simulation, it is assumed that the channel is a Rayleigh channel with zero mean, unit variance, and independent Gaussian complex random variables. The signal-to-noise ratio is SNR = P / σ 2 , where P is the user's useful signal receiving power, σ 2 is the noise power. During the simulation, the modulation mode is adjusted to make the cell rates in the two schemes approximately the same to compare the BER. Some simulation parameters are given in Table 1.

[0077] Table 1

[0078]

[0079] The present invention can eliminate cross-link interference in a dynamic TDD system and has good BER performance. It can reduce overhead through channel prediction. The present invention can also be applied to the situation of multiple cells.

[0080] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. A method for eliminating cross-link interference by RIS under dynamic TDD, characterized in that: The following steps are involved: Step 1: The transmitter forms a transmission signal; suppose the number of base station antennas is M, the number of user antennas is 1, the signal sent by the uplink cell user is x, and the signal sent by the downlink cell base station is y. Let each antenna of the base station send the same signal, that is, y = 1 M y, where 1 M represents an M×1 all-1 vector, y is the symbol sent by each antenna; Step 2: Determine the minimum number of RIS components required; the number of RIS components N must satisfy: Where BS1 represents base station 1, BS2 represents base station 2, UE1 represents user 1, UE2 represents user 2, and d BS1-BS2 represents the distance between base station 1 and base station 2, d BS1-RIS represents the distance between base station 1 and RIS, d BS2-RIS represents the distance between base station 2 and RIS, d UE1-UE2 represents the distance between user 1 and user 2, d UE1-RIS represents the distance between user 1 and RIS, d UE2-RIS represents the distance between user 2 and RIS, and the corresponding α represents the path loss exponent; Step 3: The interference stack in the dynamic TDD system is represented as B, where each element of the vector B represents the interference received by the corresponding antenna; Obtain small-scale fading matrix information, where the small-scale fading matrix information of the user and RIS is obtained by prediction due to the large acquisition overhead; define the equivalent channel matrix H according to the obtained small-scale fading matrix information D , H D The physical meaning of the element in the mth row and nth column is the sum of all channel gains passing through the nth element of RIS to the mth receiving antenna; Step 4: Reconstruct the channel matrix H D The reflection coefficient of RIS is determined by the interference B to the system Reflection coefficient of RIS in represents the amplitude coefficient of the nth element, Represents the phase shift of the nth element; the reflection coefficient of RIS It is expressed as: in represents the pseudo-inverse matrix.

2. The method for eliminating cross-link interference by RIS under dynamic TDD according to claim 1, characterized in that: The vector in step 3 is represented as: in is the small-scale fading matrix between base station 1 and base station 2; L BS1-BS2 is the path loss from BS1 to BS2, α BS1-BS2 is the path loss index between BS1 and BS2; is the small-scale fading matrix between UE1 and UE2; L UE1-UE2 is the path loss from UE1 to UE2, 3. The method for eliminating cross-link interference by RIS under dynamic TDD according to claim 2, characterized in that: Step 3 H D It is expressed as: in is the matrix H BS1-RIS The (i, j)th element of H BS1-RIS is the small-scale fading matrix between BS1 and RIS; is the matrix H BS2-RIS The (i, j)th element of H BS2-RIS is the small-scale fading matrix between BS2 and RIS; is the matrix H UE1-RIS The i-th element of H UE1-RIS is the small-scale fading matrix between UE1 and RIS; is the matrix H UE2-RIS The i-th element of H UE2-RIS is the small-scale fading matrix between UE2 and RIS; H D is a (M+1)×N matrix; H D The physical meaning of the element in the mth row and nth column of is the sum of all channel gains passing through the nth element of RIS to the mth receiving antenna.

Images